提出了一种基于小波分解和多分辨率分析的高温图像快速识别方法。首先通过小波分解降低原始图像直方图的灰度级,从而减小了分割阈值的搜索范围,再结合最大类间方差法计算分割阈值,这样可以快速找到分割阂值。实验结果表明,该方法能够显著提高分割速度并准确分割出目标图像。
This paper proposed a fast recognition method of high temperature melts based on wavelet decomposition and multiresolution analysis. Firstly, reduced the grey level of the original histogram by using the wavelet decomposition, and so was the search area of the segmentation threshold. Then found the segmentation threshold by applying Otsu' s algorithm. Experimental results show that the method can recognize the object accurately and quickly.